Electrophysiological and Proarrhythmic Effects of Hydroxychloroquine Challenge in Guinea-Pig Hearts

Dissimilar effects of stereoisomers and racemic hydroxychloroquine on Ca2+ oscillations in human induced pluripotent stem cell-derived cardiomyocytes

All currently employed pharmaceutical formulations of hydroxychloroquine (HCQ) sulfate are a racemate, consisting of equal parts mixture of two stereoisomers: R(-)HCQ and S(+)HCQ sulfates. The aims of the current study were first, to obtain and characterize pure HCQ enantiomers. The separation and purification of free base HCQ enantiomers from the racemate form were performed using semi-preparative chiral high-performance liquid chromatography. Second, we compared the pharmacological properties of both optical isomers and racemic mixture on the intracellular Ca2+ oscillations employing an in vitro model of human cardiomyocytes derived from induced pluripotent stem cells (iPSCs). The results of the pharmacological investigations indicate that the racemic and pure stereoisomer forms of HCQ sulfate exhibited a dose-dependent inhibition of spontaneous Ca2+ oscillations (as measured from their frequency and Ca2+ peak widths) in cardiomyocytes 5-45 min following exposure. In addition, the concentration-response relationships for all three compounds indicated that the rank order of potency (IC50 ) was R(-)HCQ >racemic HCQ >S(+)HCQ for the frequency of the Ca2+ oscillations and width of Ca2+ peaks for all time points examined. These studies indicate that both R(-) and S(+) stereoisomers exhibit differing pharmacological actions on hiPSC cardiomyocytes, with the former effecting a greater potency on cell Ca2+ handling.

Hydroxychloroquine and azithromycin alter the contractility of living porcine heart slices

The cardiotoxicity risk of hydroxychloroquine (HCQ) and azithromycin (AZM) has been the subject of intensive research triggered by safety concerns in COVID-19 patients. HCQ and AZM have been associated with QT interval prolongation and drug-induced arrhythmias, however other cardiotoxicity mechanisms remain largely unexplored. Our group has pioneered the living heart slice preparation, an ex-vivo platform that maintains native cardiac tissue architecture and physiological electrical and contractile properties. Here, we evaluated the cardiotoxic effect of HCQ and AZM applied alone or in combination on cardiac contractility by measuring contractile force and contraction kinetics in heart slices prepared from porcine hearts. Our results show that clinically relevant concentrations of HCQ monotherapy (1-10 µM) reduced contractile force and contraction kinetics in porcine slices in a dose-dependent manner. However, AZM monotherapy decreased contractile force and contraction kinetics only at higher concentrations (30 µM). Combination of HCQ and AZM induced a dose-dependent effect similar to HCQ alone. Furthermore, pre-treating porcine heart slices with the L-type calcium channel agonist Bay K8644 prevented the effect of both drugs, while administration of Bay K8644 after drugs interventions largely reversed the effects, suggesting a mechanism involving inhibition of L-type calcium channels. These findings indicate that HCQ and AZM alter cardiac function beyond QT prolongation with significant contractile dysfunction in intact cardiac tissue. Our porcine heart slices provide a powerful platform to investigate mechanisms of drug cardiotoxicity.

Modulation of post-pacing action potential duration and contractile responses on ventricular arrhythmogenesis in chloroquine-induced long QT syndrome

BACKGROUND

Excitation-contraction (E-C) coupling, the interaction of action potential duration (APD) and contractility, plays an essential role in arrhythmogenesis. We aimed to investigate the arrhythmogenic role of E-C coupling in the right ventricular outflow tract (RVOT) in the chloroquine-induced long QT syndrome.

METHODS

Conventional microelectrodes were used to record electrical and mechanical activity simultaneously under electrical pacing (cycle lengths from 1000-100 ms) in rabbit RVOT tissue preparations before and after chloroquine with and without azithromycin. KB-R7943 (a Na⁺-Ca²⁺ exchanger [NCX] inhibitor), ranolazine (a late sodium current inhibitor), or MgSO₄ were used to assess their pharmacological responses in the chloroquine-induced long QT syndrome.

RESULTS

Sequential infusion of chloroquine and chloroquine plus azithromycin triggered ventricular tachycardia (VT) (33.7%) after rapid pacing compared to baseline (6.7%, p = 0.004). There were greater post-pacing increases of the first occurrence of contractility (ΔContractility) in the VT group (VT vs. non-VT: 521.2 ± 50.5% vs. 306.5 ± 26.8%, p < 0.001). There was no difference in the first occurrence of action potential at 90% repolarization (ΔAPD₉₀) (VT vs. non-VT: 49.7 ± 7.4 ms vs. 51.8 ± 13.1 ms, p = 0.914). Pacing-induced VT could be suppressed to baseline levels by KB-R7943 or MgSO₄. Ranolazine did not suppress pacing-induced VT in chloroquine-treated RVOT. ΔContractility was reduced by KB-R7943 and MgSO₄, but not by ranolazine.

CONCLUSION

ΔContractility (but not ΔAPD) played a crucial role in the genesis of pacing-induced VT in the long QT tissue model, which can be modulated by NCX (but not late sodium current) inhibition or MgSO₄.

Vicious LQT induced by a combination of factors different from hERG inhibition

Clinically, drug-induced torsades de pointes (TdP) are rare events, whereas the reduction of the human ether-à-go-go-related gene (hERG) current is common. In this study, we aimed to explore the specific factors that contribute to the deterioration of hERG inhibition into malignant ventricular arrhythmias. Cisapride, a drug removed from the market because it caused long QT (LQT) syndrome and torsade de pointes (TdP), was used to induce hERG inhibition. The effects of cisapride on the hERG current were evaluated using a whole-cell patch clamp. Based on the dose-response curve of cisapride, models of its effects at different doses (10, 100, and 1,000 nM) on guinea pig heart in vitro were established. The effects of cisapride on electrocardiogram (ECG) signals and QT interval changes in the guinea pigs were then comprehensively evaluated by multi-channel electrical mapping and high-resolution fluorescence mapping, and changes in the action potential were simultaneously detected. Cisapride dose-dependently inhibited the hERG current with a half inhibitory concentration (IC50) of 32.63 ± 3.71 nM. The complete hERG suppression by a high dose of cisapride (1,000 nM) prolonged the action potential duration (APD), but not early after depolarizations (EADs) and TdP occurred. With 1 μM cisapride and lower Mg2+/K+, the APD exhibited triangulation, dispersion, and instability. VT was induced in two of 12 guinea pig hearts. Furthermore, the combined administration of isoproterenol was not therapeutic and increased susceptibility to ventricular fibrillation (VF) development. hERG inhibition alone led to QT and ERP prolongation and exerted an anti-arrhythmic effect. However, after the combination with low concentrations of magnesium and potassium, the prolonged action potential became unstable, triangular, and dispersed, and VT was easy to induce. The combination of catecholamines shortened the APD, but triangulation and dispersion still existed. At this time, VF was easily induced and sustained.